From 10884603d88eb262e314d4fbe1a7a146cc66c6d1 Mon Sep 17 00:00:00 2001 From: Jakob Unterwurzacher Date: Sat, 26 Nov 2016 12:36:55 +0100 Subject: [PATCH] benchmark.bash: double write length Writing 1000 128KB blocks takes only 1 second and yielded inconsistent results. With 2000, things look saner. --- Documentation/performance.txt | 7 +++---- tests/canonical-benchmarks.bash | 2 +- 2 files changed, 4 insertions(+), 5 deletions(-) diff --git a/Documentation/performance.txt b/Documentation/performance.txt index ac526cd..db739ec 100644 --- a/Documentation/performance.txt +++ b/Documentation/performance.txt @@ -1,10 +1,9 @@ All test performed on tmpfs, /tmp/a mounted on /tmp/b. -The used test archive is https://www.kernel.org/pub/linux/kernel/v3.0/linux-3.0.tar.gz -(mirror: http://ftp.linux.org.uk/pub/linux/linux-3.0/linux-3.0.tar.gz). +The untar test uses https://cdn.kernel.org/pub/linux/kernel/v3.0/linux-3.0.tar.gz . The archive is placed on tmpfs as well. -WRITE: dd if=/dev/zero of=zero bs=128K count=1000 -UNTAR: time tar xfz ../linux-3.0.tar.gz +WRITE: dd if=/dev/zero of=zero bs=128K count=2000 +UNTAR: time tar xzf ../linux-3.0.tar.gz LS: time ls -lR linux-3.0 > /dev/null RM: time rm -Rf linux-3.0 diff --git a/tests/canonical-benchmarks.bash b/tests/canonical-benchmarks.bash index 6dd8b4f..2e3970b 100755 --- a/tests/canonical-benchmarks.bash +++ b/tests/canonical-benchmarks.bash @@ -28,7 +28,7 @@ function etime { } echo -n "WRITE: " -dd if=/dev/zero of=zero bs=128K count=1000 2>&1 | tail -n 1 +dd if=/dev/zero of=zero bs=128K count=2000 2>&1 | tail -n 1 rm zero sleep 1 echo -n "UNTAR: "